A Fractal Multi-wavelets Filter Design and Application

In this paper, in order to obtain the more accurate local features of the iris image, firstly, on the basis of the first-order Taylor expansion of Gauss Function, orthonormality is employed to obtain a set of complete quadratic units orthogonal basis, secondly, these quadratic units orthogonal basis are used as the scaling function, and a biorthogonal multi-wavelets filter is designed. This filter has many characteristics, such as symmetry, compact support, orthogonality and 3-order vanishing moment, lastly, the algorithms of decomposition and reconstruction of this filter and the test about the decomposition and reconstruction of iris image are given. The results of the experiment and the analysis show that by decomposing the filter matrix. 36 direction features of an image can be obtained and this method is a good feature extraction one.